Macrophage targeted theranostic strategy for accurate detection and rapid stabilization of the inflamed high-risk plaque

Rationale: Inflammation plays a pivotal role in the pathogenesis of the acute coronary syndrome. Detecting plaques with high inflammatory activity and specifically treating those lesions can be crucial to prevent life-threatening cardiovascular events. Methods: Here, we developed a macrophage mannose receptor (MMR)-targeted theranostic nanodrug (mannose-polyethylene glycol-glycol chitosan-deoxycholic acid-cyanine 7-lobeglitazone; MMR-Lobe-Cy) designed to identify inflammatory activity as well as to deliver peroxisome proliferator-activated gamma (PPARγ) agonist, lobeglitazone, specifically to high-risk plaques based on the high mannose receptor specificity. The MMR-Lobe-Cy was intravenously injected into balloon-injured atheromatous rabbits and serial in vivo optical coherence tomography (OCT)-near-infrared fluorescence (NIRF) structural-molecular imaging was performed. Results: One week after MMR-Lobe-Cy administration, the inflammatory NIRF signals in the plaques notably decreased compared to the baseline whereas the signals in saline controls even increased over time. In accordance with in vivo imaging findings, ex vivo NIRF signals on fluorescence reflectance imaging (FRI) and plaque inflammation by immunostainings significantly decreased compared to oral lobeglitazone group or saline controls. The anti-inflammatory effect of MMR-Lobe-Cy was mediated by inhibition of TLR4/NF-κB pathway. Furthermore, acute resolution of inflammation altered the inflamed plaque into a stable phenotype with less macrophages and collagen-rich matrix. Conclusion: Macrophage targeted PPARγ activator labeled with NIRF rapidly stabilized the inflamed plaques in coronary sized artery, which could be quantitatively assessed using intravascular OCT-NIRF imaging. This novel theranostic approach provides a promising theranostic strategy for high-risk coronary plaques.


Determination of amount of lobeglitazone and Cy7 within the theranostic nanodrug
To determine drug loading content of lobeglitazone in MMR-Lobe-Cy, MMR-Lobe-Cy (1 mg) was solubilized in 1 mL of acetonitrile:deionized water:formic acid (60:40:0.25, v/v/v), and highperformance liquid chromatography (HPLC; Agilent 1260 series, Agilent Technologies, Santa Clara, CA, USA) was performed [1,2]. In brief, 5 μL of MMR-Lobe-Cy (1 mg/mL) was eluted at a flow rate of 0.5 mL/min via a mobile phase (acetonitrile:deionized water:formic acid (60:40:0.25, v/v/v)). The calculated amount of lobeglitazone in 1 mg of MMR-Lobe-Cy was 0.258 mg by using the equation obtained from the standard curve of the lobeglitazone [Y = 19446X + 0.7106: R 2 = 1]. To determine the amount of Cy7, MMR-Lobe-Cy (1 mg) and MMR-Cy (1 mg) were clearly dissolved in dimethyl sulfoxide (DMSO; 1 mL) and the absorbance was measured at 760 nm using a UV-VIS spectrophotometer (NEOGEN, Daejeon, South Korea). The calculated amount of Cy7 in 1 mg of MMR-Lobe-Cy and MMR-Cy was 0.75 μg and 1.6 μg, respectively, by using the following equation obtained from the standard curve of the Cy7 [Y = 359.95X + 0.0319

Particle size analysis
For particle size analysis, MMR-Lobe-Cy (1 mg) was dispersed in deionized water (1 mL) and then sonicated for 1 min at 100 W. The particle sizes were determined using a Zetasizer 3000 instrument (Malvern Instruments, Malvern, UK).

Transmission electron microscopy
To determine the shape of the nanoparticle, MMR-Lobe-Cy (0.5 mg) was dispersed in 1 mL of deionized water by vortexing for 1 min, placed on grid, negatively stained with 2 wt% uranylacetate solution, and let to dry. The morphology of MMR-Lobe-Cy nanoparticle was observed using an energy filtering transmission electron microscopy (EF-TEM; LEO 912AB OMEGA, Carl Zeiss, Oberkochem, Germany).

In vitro drug release study
To evaluate the release profile of lobeglitazone from MMR-Lobe-Cy, we dispersed the 1 mg of lyophilized MMR-Lobe-Cy in 1 mL of PBS (pH 7.4) by vortexing and sonification (100 W) for 1 min, respectively. Then, the dispersed MMR-Lobe-Cy was introduced into a dialysis membrane (MWCO 6-9 kDa) which was immersed in PBS (20 mL, pH 7.4), and then shaking and oscillating 100 times/min in a water bath (37℃). The amount of lobeglitazone released from MMR-Lobe-Cy was measured by HPLC at predetermined time points, followed by medium replacement with fresh one at each time. We used acetonitrile:H2O:formic acid (60:40:0.25) as a mobile phase. The analysis was carried out at 0.5 mL/min flow rate with a detection wavelength of 250 nm and the injection volume was arranged as 5 μL.

The stability test of MMR-Lobe-Cy in PBS and 10% FBS-containing DMEM
To test the stability of MMR-Lobe-Cy, 0.5 mg of MMR-Lobe-Cy was added to PBS (pH 7.4) and 10% FBS-containing DMEM (without phenol) and then sonicated at 22.5 W for 1 min using a probe-type sonicator. Then, the hydrodynamic diameters of MMR-Lobe-Cy were measured in two different conditions for 6 days.

Quantitative PCR
Total RNA was extracted from RAW264.7 cells with AccuPrep® Universal RNA Extraction Kit (K-3141; Bioneer, Daejeon, Korea) according to the manufacturer's instructions. One microgram total RNA for each sample was reverse-transcribed using the AccuPower® RocketScript Cycle RT premix (K-2201; Bioneer). The resulting complementary DNA was analyzed by real-time PCR using AccuPower® 2X GreenStar™ qPCR MasterMix (K-6251; Bioneer) under the QuantStudio 6 Flex Real-time PCR system (Thermo Fisher Scientific, Waltham, MA, USA). The cycling conditions were 95℃ for 10 min, followed by 40 cycles of 95℃ for 5 sec, 58℃ for 25 sec and 72℃ for 30 sec. Primers that were used are listed in Table S1. Relative mRNA levels were determined using the comparative Ct method and normalized against GAPDH. Immunofluorescence RAW264.7 cells (1×10 5 cells per plate) were seeded in poly L-lysine-coated chamber slides, incubated for 24 h, and then stimulated with LPS and LDL for 24 h. The cells were fixed with 4% paraformaldehyde in PBS for 20 min at room temperature, blocked with 2% bovine serum albumin for 1 h, and incubated with CD206 antibody (MR5D3; 1:20; Bio-Rad Laboratories, Hercules, CA, USA) overnight at 4 C, followed by incubation with AffiniPure goat anti-rat IgG secondary antibody (405418; 1:50) overnight at 4℃. After washing with PBS, the cells were mounted using a fluorescence mounting medium with DAPI (#E19-18; GBI Labs, Bothell, WA, USA) and examined using a confocal fluorescence microscope (LSM 900, Carl Zeiss, Oberkochen, Germany).

Development of Atheromatous Rabbit Model
New Zealand white rabbits (male, 3-month-old; Doo-Yoel Biotech, Seoul, South Korea) served as the atheromatous rabbit model. Atherosclerotic lesions were developed by denuding aortic endothelium with balloon catheter and inducing hypercholesterolemia with high cholesterol dietfeeding (Doo-Yeol Biotech, Seoul, South Korea). New Zealand white rabbits were fed high cholesterol diet (1% cholesterol) for 1 week before balloon injury to induce hypercholesterolemia.
3F Fogarty embolectomy catheter (Edwards Laboratories, Santa Ana, CA, USA) was inserted through carotid artery and balloon denudation was performed at the infrarenal aorta with three pullbacks at a balloon pressure ranging from 0.15 to 0.2 mL to distend the aorta to 1.5-fold of the diameter. Three weeks after balloon injury, rabbits were placed on the 0.1% high cholesterol diet for 6-8 weeks, and normal diet was given with the start of the treatment. After 4 weeks of 1% HCD feeding, the amount of dietary cholesterol supplementation was reduced to 0.1% cholesterol and maintained until the in vivo experiments to prevent hepatotoxicity, since the long-term administration of high-cholesterol diet could result in overt fatty liver disease [3].

Catheter-based intravascular OCT-NIRF imaging system
We have developed a custom-built catheter-based OCT-NIRF imaging system. Our OCT system obtains high-resolution (axial resolution of ~13 μm) OCT images at an A-scan rate of 117.2 kAlines/sec and a high frame rate of up to 100 frames/sec using a custom-built high-speed wavelength-swept laser with a repetition rate of 117.2 kHz and a wavelength sweeping range of 110 nm (centered at 1,295 nm) [7]. Excitation of Cy7 dye was done via a 730 nm laser diode

Distance calibration and normalization of NIRF emission intensity
Since the NIRF emission intensity attenuates as a function of the distance from the imaging catheter to the arterial wall [8], a calibration algorithm for the attenuated intensity is required for accurate and undistorted inflammation assessment based on NIRF data. Details for the calibration algorithm has been described in our previous publications [4,5]. Briefly, first, OCT-NIRF imaging was performed on a fluorescent phantom tube filled with homogeneous Cy7 milk solution to obtain the relationship between the NIRF emission intensity and the distance. Second, the distance between the imaging catheter and the surface of the phantom tube was estimated in all OCT images of the pullback by using the automated lumen segmentation algorithm [9]. Third, the pairs between the NIRF emission intensity and the corresponding distance were approximated by a two-term exponential fitting function: Despite the calibration for the attenuation along the distance, the NIRF emission intensity may still be affected by other factors, such as variations of the imaging system (imaging catheter, laser diode output, and detector sensitivity) and differences in pharmaco-kinetics and -dynamics between individual rabbits. In order to exclude the potential distortion of the NIRF emission intensity due to the factors mentioned above, we normalized the NIRF emission intensity to the plaque target-to-background ratio (pTBR): where and ̅̅̅̅̅̅̅ denote the distance-calibrated NIRF emission intensity and its background value, respectively, and (•) denotes the operators for pTBR calculation. The background value was determined by averaging the five lowest NIRF emission intensities in neighboring normal-looking intact segments. On the graphical user interface for OCT-NIRF image acquisition, the entire calibration and normalization processes were automatically performed as post-processing.

Picro-Sirius Red staining
For Picro-Sirius Red (PSR) Staining, the sections were hydrated to distilled water and placed in phosphomolybdic acid solution (0.2%) for 10 sec, followed by dipping slide in distilled water.
Next, the slides were incubated with PSR solutions for 1h, and then rinsed quickly two times in acetic acid solution (0.5%). The sections were dehydrated, cleared and mounted in synthetic resin.

Detailed methods for quantifying Oil-red-O and Picro-Sirius Red stained images
Oil-red O (ORO) and/or PSR-stained tissue images were properly adjusted by white balance.
Second, to separate the red positive-stained pixels from both ORO and PSR staining images, the default red-green-blue (RGB) color-space was converted to hue-saturation-value (HSV) colorspace. Third, Gaussian filtering with a sigma of 20 pixels, binarization by thresholding and image erosion by a structuring element with a radius of 25 pixels were sequentially performed on the value channel for the segmentation of the tissue section areas. Note that since background of the images obtained by brightfield microscopy was white, the value channel was inverted to set the background as zero before the following image processing. Fourth, after applying Gaussian filtering with a sigma of 20 pixels to the saturation channel, red positive-stained pixels in the predetermined tissue section areas were chosen according to the corresponding hue (-50º ~ 50º) and saturation (>6%) channels. Finally, the quantified positive-stained ratio was calculated by dividing the number of pixels in the segmented tissue section area by that in the positive-stained area. The parameters introduced above have been determined empirically and can be adjusted slightly depending on the type of the histological staining. This quantitation processes were implemented using ImageJ (National Institute of Health, Bethesda, MD, USA) and MATLAB software (R2017a; The MathWorks, Natick, MA, USA).

Complete blood count and biochemistry
Blood was collected in ethylenediaminetetraacetic acid (EDTA) tube, and then hematological parameters such red blood cells, hemoglobin, platelets, and white blood cells were analyzed using an automated hematology system (XN-9100; Sysmex Corporation, Kobe, Japan). To measure the biochemistry profile, blood was sampled in serum separator blood collection tube, left to clot for at least 30 min, and subsequently centrifuged at 3000 rpm for 5 min. The serum was aliquoted and stored at -80℃ until measured on an automatic analyzer (DRI-CHEM NX500i; FUJIFLIM, Japan).